| { | |
| "best_metric": 0.7773279352226721, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-vrcn3dxi/checkpoint-300", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.9803097248077393, | |
| "learning_rate": 5.0385074972904444e-05, | |
| "loss": 0.6147, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.1125147342681885, | |
| "learning_rate": 4.9356808136722715e-05, | |
| "loss": 0.4868, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.646878719329834, | |
| "learning_rate": 4.832854130054099e-05, | |
| "loss": 0.4208, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.7291347980499268, | |
| "learning_rate": 4.730027446435928e-05, | |
| "loss": 0.4988, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 4.168182849884033, | |
| "learning_rate": 4.627200762817755e-05, | |
| "loss": 0.3829, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 5.242916584014893, | |
| "learning_rate": 4.524374079199583e-05, | |
| "loss": 0.371, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 6.568434715270996, | |
| "learning_rate": 4.42154739558141e-05, | |
| "loss": 0.3587, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 6.341504096984863, | |
| "learning_rate": 4.318720711963238e-05, | |
| "loss": 0.2841, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 6.951382160186768, | |
| "learning_rate": 4.2158940283450655e-05, | |
| "loss": 0.5014, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.481485605239868, | |
| "learning_rate": 4.113067344726893e-05, | |
| "loss": 0.3564, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.825, | |
| "eval_f1": 0.75177304964539, | |
| "eval_loss": 0.38982895016670227, | |
| "eval_precision": 0.6666666666666666, | |
| "eval_recall": 0.8617886178861789, | |
| "eval_runtime": 1.5252, | |
| "eval_samples_per_second": 262.256, | |
| "eval_steps_per_second": 16.391, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.624406099319458, | |
| "learning_rate": 4.010240661108721e-05, | |
| "loss": 0.3017, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 11.477983474731445, | |
| "learning_rate": 3.907413977490549e-05, | |
| "loss": 0.414, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 3.667001724243164, | |
| "learning_rate": 3.804587293872376e-05, | |
| "loss": 0.4467, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.0119214057922363, | |
| "learning_rate": 3.701760610254204e-05, | |
| "loss": 0.2403, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 4.618969440460205, | |
| "learning_rate": 3.5989339266360316e-05, | |
| "loss": 0.2895, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 6.767045497894287, | |
| "learning_rate": 3.4961072430178594e-05, | |
| "loss": 0.2201, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 4.485604286193848, | |
| "learning_rate": 3.393280559399687e-05, | |
| "loss": 0.3093, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 4.330203056335449, | |
| "learning_rate": 3.290453875781515e-05, | |
| "loss": 0.2316, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.8137776851654053, | |
| "learning_rate": 3.187627192163342e-05, | |
| "loss": 0.2012, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.9346457719802856, | |
| "learning_rate": 3.08480050854517e-05, | |
| "loss": 0.2902, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8525, | |
| "eval_f1": 0.7686274509803922, | |
| "eval_loss": 0.34935688972473145, | |
| "eval_precision": 0.7424242424242424, | |
| "eval_recall": 0.7967479674796748, | |
| "eval_runtime": 1.5077, | |
| "eval_samples_per_second": 265.312, | |
| "eval_steps_per_second": 16.582, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 2.6348159313201904, | |
| "learning_rate": 2.9819738249269974e-05, | |
| "loss": 0.144, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 6.430622100830078, | |
| "learning_rate": 2.8791471413088256e-05, | |
| "loss": 0.1821, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 4.538208484649658, | |
| "learning_rate": 2.776320457690653e-05, | |
| "loss": 0.3038, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 3.8831608295440674, | |
| "learning_rate": 2.673493774072481e-05, | |
| "loss": 0.1824, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.0167384147644043, | |
| "learning_rate": 2.5706670904543083e-05, | |
| "loss": 0.1242, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 1.3187135457992554, | |
| "learning_rate": 2.4678404068361358e-05, | |
| "loss": 0.145, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 6.20335054397583, | |
| "learning_rate": 2.365013723217964e-05, | |
| "loss": 0.2031, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 15.89792251586914, | |
| "learning_rate": 2.2621870395997914e-05, | |
| "loss": 0.1694, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 8.923992156982422, | |
| "learning_rate": 2.159360355981619e-05, | |
| "loss": 0.1949, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.10263983905315399, | |
| "learning_rate": 2.0565336723634466e-05, | |
| "loss": 0.0461, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8625, | |
| "eval_f1": 0.7773279352226721, | |
| "eval_loss": 0.46338924765586853, | |
| "eval_precision": 0.7741935483870968, | |
| "eval_recall": 0.7804878048780488, | |
| "eval_runtime": 1.5254, | |
| "eval_samples_per_second": 262.232, | |
| "eval_steps_per_second": 16.39, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 635446111352832.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 5.1413341809086166e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 16, | |
| "seed": 1 | |
| } | |
| } | |